Assessment of cue counting for estimating bird density using passive acoustic monitoring: recommendations for estimating a reliable cue rate
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
VOLUME 16, ISSUE 1, ARTICLE 11 Pérez-Granados, C., A. Barrero, J. Traba, D. Bustillo-de la Rosa, M. Reverter, and J. Gómez-Catasús. 2021. Assessment of cue counting for estimating bird density using passive acoustic monitoring: recommendations for estimating a reliable cue rate. Avian Conservation and Ecology 16 (1):11. https://doi.org/10.5751/ACE-01801-160111 Copyright © 2021 by the author(s). Published here under license by the Resilience Alliance. Methodology Assessment of cue counting for estimating bird density using passive acoustic monitoring: recommendations for estimating a reliable cue rate Cristian Pérez-Granados 1,2, Adrián Barrero 1,3, Juan Traba 1,3, Daniel Bustillo-de la Rosa 1,3, Margarita Reverter 1,3 and Julia Gómez- Catasús 1,3 1 Terrestrial Ecology Group (TEG-UAM), Department of Ecology, Universidad Autónoma de Madrid, Madrid, Spain, 2Ecology Department/IMEM "Ramón Margalef", Universidad de Alicante, Alicante, Spain, 3Centro de Investigación en Biodiversidad y Cambio Global (CIBC-UAM), Universidad Autónoma de Madrid, Madrid, Spain ABSTRACT. Cue counting is a method developed for estimating vocally active wildlife density by dividing the density of cues (number of cues per unit area surveyed per unit time) by the average cue rate (ACR) at which individuals vocalize. It has been used successfully to estimate whale density using passive acoustic monitoring, but its efficacy has had limited testing in birds. We tested whether cue counting can be used to infer bird abundance using autonomous recording units and estimated the minimum effort required to obtain a reliable cue rate at individual and population levels. We recorded Dupont's Lark (Chersophilus duponti) vocalizations at 31 sites where traditional field censuses were also performed. We estimated the ACR using three methodologies: directional recordings, recordings from an online database of bird sounds (xeno-canto), and behavioral field studies. The ACRs estimated using directional recordings and behavioral field studies were similar, and bird numbers were over and underestimated by 0.8 and 10%, respectively (74–77% of the sampling sites were well estimated). However, the ACR estimated using xeno-canto recordings was much higher than those estimated using the other two methods, and bird numbers were underestimated by 41%. We also performed a cost-effectiveness assessment of the number of individuals and recording durations needed to optimize the estimation of a reliable ACR. We found that ACR estimates were more efficient if long (25 min) recordings were used when < 4 males were recorded, whereas 5-min recordings were more efficient for ≥ 20 males. We conclude that cue counting can be useful to infer bird density around recorders but requires an accurate measure of the ACR. Further research should evaluate the effectiveness of passive cue counting on a large number of species and under different circumstances. Évaluation du comptage des détections pour estimer la densité d'oiseaux à l'aide d'un suivi sonore passif : recommandations pour estimer un taux de détections fiable RÉSUMÉ. Le comptage des détections est une méthode qui a été élaborée pour estimer la densité de la faune active vocalement en divisant la densité de détections (nombre de détections par unité de surface étudiée par unité de temps) par le taux moyen de détections (TMD) auquel les individus chantent ou crient. Cette méthode a été utilisée avec succès pour calculer la densité des baleines à l'aide d'un suivi sonore passif, mais son efficacité a peu été testée chez les oiseaux. Nous avons testé si le comptage des détections pouvait être utilisé pour déduire l'abondance des oiseaux en utilisant des enregistreurs automatisés et avons calculé l'effort minimum requis pour obtenir un taux de détections fiable au niveau des individus et des populations. Nous avons enregistré les manifestations sonores du Sirli de Dupont (Chersophilus duponti) sur 31 sites où des recensements traditionnels ont également été effectués. Nous avons calculé le TMD de trois façons : à partir d'enregistrements directionnels, d'enregistrements provenant d'une base de données de cris et de chants d'oiseaux accessible en ligne (xeno-canto) et d'études comportementales sur le terrain. Les TMD calculés à l'aide d'enregistrements directionnels et d'études comportementales étaient similaires, et le nombre d'oiseaux était surestimé et sous-estimé de 0,8 et 10 %, respectivement (la mesure de 74-77 % des sites d'échantillonnage était juste). Cependant, le TMD calculé à l'aide d'enregistrements de xeno-canto était beaucoup plus élevé que ceux obtenus au moyen des deux autres méthodes, et le nombre d'oiseaux était sous-estimé de 41 %. Nous avons également réalisé une évaluation coût-efficacité du nombre d'individus et des durées d'enregistrement nécessaires pour optimiser le calcul d'un TMD fiable. Nous avons constaté que la mesure du TMD était meilleure si des enregistrements longs (25 min) étaient utilisés lorsque < 4 mâles étaient enregistrés, tandis que des enregistrements de 5 min étaient plus efficaces pour ≥ 20 mâles. Nous concluons que le comptage des détections peut être utile pour calculer la densité d'oiseaux autour des enregistreurs, mais une mesure précise du TMD doit d'abord être effectuée. D'autres recherches devraient se pencher sur l'évaluation de l'efficacité du comptage passif des détections dans le cas d'un grand nombre d'espèces et dans différentes circonstances. Key Words: autonomous recording unit; Chersophilus duponti; population estimates; vocal activity; wildlife monitoring; xeno-canto Address of Correspondent: Cristian Pérez-Granados, Terrestrial Ecology Group (TEG-UAM), Department of Ecology, Universidad Autónoma de Madrid, Madrid, Spain, Tel: +34 (914978271), Fax: +34 (914978001), cristian.perez@ua.es
Avian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
INTRODUCTION detection radius for passive acoustic monitoring; Buckland et al.
The use of autonomous recording units (ARUs) for wildlife 2001). Buckland (2006) used cue counts heard from point counts
monitoring has grown considerably in recent years (Shonfield and to estimate the density of four passerine birds and concluded that
Bayne 2017, Sugai et al. 2019). Bird surveys are usually based on cue counting might be a workable method for specialist surveys
auditory cues, and therefore, birds are among the taxa most of single species. Later, Buckland et al. (2008) stated that cue
commonly surveyed using ARUs (Sugai et al. 2019). Although counting has potential for estimating bird abundance using
ARUs have been used in many avian research areas, their use acoustic arrays. In the first assessment using the cue rate of a bird
during the early years of the technique was mainly restricted to species and ARUs, Lambert and McDonald (2014) applied cue
estimating species richness and detecting the presence of target counting to estimate Bell Miner (Manorina melanophrys) density
species at specific sites (Shonfield and Bayne 2017). Along with within a fixed radius of ARUs and concluded that acoustic
methodological and technical advances, low-cost ARUs have surveys could detect more individuals than could traditional field
become available recently (e.g., Hill et al. 2018, Wijers et al. 2021), surveys. Subsequently, Sebastián-González et al. (2018)
opening new avenues for further progress in the use of ARUs for developed a protocol to estimate bird density using the target
bird monitoring. species’ cue rate and including covariates such as environmental
conditions and an estimate of the distance of vocalizing
Among the technical advances, it is worth highlighting an increase individuals to the recorder. They showed that cue rate, when
in computational capabilities and the development of effective coupled with other covariates, is an effective method to estimate
and user-friendly automated signal recognition software (Knight bird density using ARUs (Sebastián-González et al. 2018).
et al. 2017, Priyadarshani et al. 2018), enabling researchers to
analyze large amounts of data in a timely manner. One of the Cue counting is a rapid method that can be applied easily,
major challenges for monitoring birds using ARUs is the difficulty estimates the number of individuals around ARUs, and is easy to
in estimating the number of birds vocalizing around recorders. interpret by unskilled bioacousticians (general public, managers,
Although many methods and statistical approaches have been and stakeholders). One of the main advantages of cue counting
described for estimating bird population density around ARUs, (active or passive) is that it may be applied without the collection
their effectiveness has been tested on only a few species (e.g., Oppel of complementary data while recording to estimate bird
et al. 2014, Drake et al. 2016, Van Wilgenburg et al. 2017, Darras abundance, in contrast to other methods such as sound-level
et al. 2018, Sebastián-González et al. 2018, Bombaci and Pejchar measurement (Yip et al. 2020), the vocal activity index (Oppel et
2019, Pérez-Granados et al. 2019c, Yip et al. 2020; reviewed by al. 2014), and paired acoustic sampling (Van Wilgenburg et al.
Pérez-Granados and Traba 2021). Some methods that are already 2017, Bombaci and Pejchar 2019). Nonetheless, prior knowledge
used for estimating mammal abundance with ARUs have scarcely about the sampling radius of the recorder is required to obtain
been tested on birds, for example, passive acoustic cue counting robust estimates when applying cue counting (Lambert and
(hereafter cue counting; Hiby and Ward 1986, Marques et al. McDonald 2014). Another advantage of cue counting is that it
2009, 2011, but see Lambert and McDonald 2014, Sebastián- requires little effort and expertise for recording interpretation
González et al. 2018). compared to other methods such as the recognition of individuals
through sonogram analyses (Ehnes and Foote 2015). However,
Cue counting relies on the idea that the number of target species an important disadvantage is that cue counting requires a reliable
vocalizations in a sound recording is related to the number of estimation of the target species’ ACR to convert number of bird
animals vocalizing around recorders (density dependent, similar vocalizations detected by ARUs to bird density. ACR estimation
to the vocal activity index; see Farnsworth et al. 2004, Oppel et may require monitoring the cue rate of several individuals over
al. 2014). Determining an unbiased average cue rate (ACR) for a large temporal scales, which might be difficult for some species
target species at the population level (vocalizations per unit time; (e.g., shy and threatened birds) and may differ among populations.
Buckland 2006) allows researchers to estimate the number of ACR can be estimated using various approaches such as by
individuals around ARUs (Lambert and McDonald 2014). The placing ARUs in areas known to be occupied by a single
ACR can be estimated by averaging the cue rate at which individual, using archived recordings in private or public sound
individual birds produce vocalizations (vocalizations per unit databases, or performing detailed observational studies.
time per bird) and must be based on a representative sample of Nonetheless, the effectiveness of estimating ACR using different
birds (Buckland et al. 2008). Cue counting was originally methods has never been assessed. Similarly, there is limited
developed for estimating whale density from recorded whale information about the number of individuals and the minimum
blows and has proven to be a useful method when it is possible to period that each individual should be monitored to estimate a
detect and count some vocal cue, but it is difficult to identify which reliable ACR.
individual produced which cue (Marques et al. 2013). In the case
of whales, density can be estimated by dividing the density of cues We had two distinct objectives for our study. First, we aimed to
(number of whale blows per unit area surveyed per unit time) by evaluate whether cue counting can reliably infer bird abundance
an estimate of the average rate at which whales vocalize (number around ARUs, using the Dupont’s Lark (Chersophilus duponti) as
of blows per unit time; Hibby and Ward 1986, Marques et al. a model species. To do so, we deployed ARUs in 31 habitat patches
2009, 2011, 2013). The method could be used for estimating bird (sampling sites) with known population size determined using
densities using ARUs if cue rates are well estimated and unbiased traditional field surveys and annotated the number of
(see Sebastián-González et al. 2018); the number of birds vocalizations recorded at each sampling site per unit time. We
vocalizing around ARUs could be estimated by dividing the total applied cue counting to estimate the number of individuals per
number of bird vocalizations detected per recording length by the sampling site by applying the ACR of the species estimated using
ACR of the target bird species, when including a measurement three approaches: (1) directional recordings previously collected
of the area efficiently surveyed (e.g., estimating the effective in the study area, (2) existing recordings uploaded to a public
Avian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
sound database, and (3) published behavioral field studies. ARU consisted of a USB Voice Recorder SK-001 equipped with
Second, we aimed to perform a cost-effectiveness assessment of a single-channel microphone and a digital timer. Devices were
ACR estimation to identify the minimum number of males and powered by a 12-V/8.0-mAh battery (> 300 h autonomy) and
minimum recording duration for Dupont’s Lark to obtain an protected from weather by small and cryptic plastic boxes (see
accurate ACR while maximizing survey effectiveness. Although Fig. S2 in Appendix 1). ARUs were programmed to make a unique
our research was focused on a single bird species, some of the 30-min recording per sampling site between 60 and 30 min before
lessons learned should apply to other vocally active taxa, sunrise (the period of maximum vocal activity of the species;
including nonavian species. Pérez-Granados et al. 2018a). Recordings always used a sample
rate of 20 kHz (range 20 Hz to 20 kHz), 16 bits in mono mode,
METHODS and the recorder had a signal-to-noise ratio of 80 dB. Audio was
recorded in mp3 format using a bitrate of 128 kbps, and recordings
Study species were stored on microSD memory cards capable of storing ~ 500
We selected Dupont’s Lark as the study species because there are h of continuous data. Recordings were collected during the
already several studies about its vocal behavior (including species’ breeding period (mid-April and mid-June) throughout
measures of ACR) and the use of ARUs for monitoring it, 2017–2019.
providing a useful background for comparison and deeper
The ARU used can record the species present up to 256 m distant
analysis (e.g., Laiolo and Tella 2005, Laiolo et al. 2007, 2008,
(Pérez-Granados et al. 2019a). However, the probability of
Pérez-Granados et al. 2016). Dupont’s Lark is a diurnal passerine
detecting a bird vocalization is related to the distance of the
with high vocal activity concentrated at dawn, when it utters a
vocalizing bird from the ARU (Yip et al. 2017, Sebastián-
large number of songs and territorial calls, its main vocalization
González et al. 2018). Therefore, we aimed to identify a threshold
types (Laiolo et al. 2018, Pérez- Granados et al. 2018b). Dupont’s
volume at which Dupont’s Lark whose vocalizations exceeded the
Lark can sing both from the ground and while flying, as do several
threshold were likely to be within the selected radius (see Hedley
other larks (de Juana et al. 2004, Gómez-Catasús et al. 2016). It
et al. 2021). To identify this threshold, we performed a playback
is assumed that only Dupont’s Lark males utter songs and
trial in the species’ typical habitat and measured the rate of
territorial calls (Pérez-Granados et al. 2018c), so we focused on
amplitude decline of broadcast vocalizations under real
vocalizing males. The Dupont’s Lark song usually comprises 4–
conditions (for a similar approach see Lambert and McDonald
5 discrete song types, which are typically shared among neighbors
2014). We broadcasted 26 different Dupont’s Lark vocalizations
and repeated in the same order (Laiolo et al. 2008, Pérez-
at each of nine definite distances between 1 and 256 m (full details
Granados et al. 2016). Calls consist of discrete and short whistles
of the playback test can be found in Pérez-Granados et al. 2019a).
(Laiolo et al. 2007), and each whistle is considered a single
For each vocalization uttered at each distance, we annotated
vocalization. The call and the last song type of the song follow a
whether it was visible on the spectrogram (Hamming Window,
common sequence, the so-called “whee-ur-wheeee” (Cramp 1988,
512-sample window length and 50% overlap, see Table S1 in
see Fig. S1 in Appendix 1). Here, we refer to the whole sequence
Appendix 1) and measured its maximum amplitude (hereafter
of song types as a song and use the definite “whee-ur-wheeee” to
amplitude) from the waveform (Lambert and McDonald 2014,
count the number of songs within a recording. This interpretation
Hedley et al. 2021) using Raven Pro 1.5 software (Center for
means that each train of songs, usually composed of 4–5 song
Conservation Bioacoustics 2014). Vocalization amplitude was
types, was considered to be one cue (Fig. S1 in Appendix 1). We
measured within the frequency bin of 2–5 kHz to minimize the
used this approach to be consistent with the methodology
influence of background and uninteresting sounds in the
proposed by Pérez-Granados et al. (2018c) and to facilitate
measurement while considering the frequency range of the target
comparisons with our acoustic data. The same methodology was
species (Stowell and Plumbley 2014, Hedley et al. 2021). Our
used to count the number of songs throughout our work.
analyses identified that Dupont’s Lark vocalizations uttered
within 160 m of the recorder would have a mean minimum
Study area amplitude of 144.5 units (U), which was distinguishable from calls
The study area was located in central (40º37’ N, 3º09’ W,
uttered from further away (Fig. 1). Moreover, the probability of
Guadalajara and Soria provinces, 28 sampling sites) and
detecting a Dupont’s Lark vocalization uttered within the 160 m
northeastern Spain (41º30’ N, 0º37’ E, Lleida province, three
radius was 0.974 (152 of 156 vocalizations detected, see Table S1
sampling sites) and comprised 31 sampling sites occupied by the
in Appendix 1). We used this threshold amplitude when analyzing
Dupont’s Lark. Sampling sites were flat areas dominated by
the recordings made at sampling sites to restrict detections to this
shrubs and scrubs < 40 cm tall such as thyme (Thymus spp.),
pre-specified radius. Thus, vocalizations with amplitude > 144.5
broom (Genista spp.), and lavender (Lavandula spp.), and a high
U were considered to be produced by males within 160 m
proportion of bare ground. Detailed information about the
(Lambert and McDonald 2014), and a probability of detection
typical habitat and climate in the three study areas are provided
of 1 was assumed. We did not calibrate our measurements to the
by Pérez-Granados et al. (2018b, 2019a) for Guadalajara and
sound pressure of the microphone, so the described relationship
Catalonia province, respectively, and Gómez-Catasús et al. (2019)
between vocalization amplitude and distance is only valid for our
for Soria province.
ARU, habitat surveyed, and species monitored. The spectrograms
of the audio recordings collected at sampling sites were also
Acoustic monitoring and recording
visually inspected using Raven Pro 1.5 (Center for Conservation
interpretation Bioacoustics 2014) by the same observer (CPG). For each
At each sampling site, we placed an ARU created ad-hoc for recording, the waveform line was shifted until it reached the
Dupont’s Lark monitoring (Pérez-Granados et al. 2019a). TheAvian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
threshold 144.5 U, and we counted the number of Dupont’s Lark The cue rate of each male was estimated as the total number of
cues (songs and territorial calls) that reached the threshold (see vocalizations detected per recording by visually inspecting the
Lambert and McDonald for a similar approach). spectrograms divided by recording length (in minutes). The
specific methods employed for each methodology were as follows.
Fig. 1. Relationship between maximum amplitude (Units; mean
± standard error) of Dupont’s Lark vocalizations and distance Directional recordings: We recorded 64 different Dupont’s Lark
to playback location at nine distances. Vocalizations uttered at males during April and May of 2016 in the study area. Recordings
1 m are excluded for graphical purposes. were made using a digital portable memory recorder (Tascam
DR-40 version 2) and a Sennheiser ME67 shotgun microphone
with a K6 powering unit and covered with a Gutmann windshield.
A single male was recorded per recording at a maximum distance
of 50 m to the observer during the dawn period. Because distances
to the observer were short and only single-male recordings were
collected, we assumed that the probability of detecting a Dupont’s
Lark vocalization on directional recordings was 1. Recording
duration varied between 302 and 489 s (mean ± standard deviation
[SD]: 337.1 ± 33.9 s).
Xeno-canto database: We used the xeno-canto online database of
sounds (https://www.xeno-canto.org/) and downloaded all
available recordings of the Dupont’s Lark (on 02 January 2020).
Thirty-nine recordings were available. Five of the recordings were
removed due to poor quality or because we were unable to monitor
the cue rate of a single individual (i.e., several males were
vocalizing in the recording). Although we could not evaluate the
distance of the recorded bird to the observer, we applied the same
reasoning as in the directional recordings and assumed that the
probability of detecting a Dupont’s Lark vocalization on
directional recordings was 1. Recording length varied between 2
Average cue rate and 217 s (mean ± SD: 78.6 ± 57.3 s). Because of low sample size
We estimated the ACR of the Dupont’s Lark using three methods. and short recording length, we considered all the recordings
uploaded to xeno-canto and not just those performed during the
Behavioral field study: In this approach, we estimated the species’ breeding season at dawn.
ACR using published data about its vocal behavior. We used data
collected by Pérez-Granados et al. (2018c), which are freely Bird data censuses
available. Pérez-Granados et al. (2018c) monitored the vocal We performed a traditional field survey at each sampling site to
behaviour of Dupont’s Lark males during the breeding season in estimate the number of Dupont’s Lark males within a 160-m
three populations located in eastern Spain over two years. For buffer of each ARU, according to the threshold radius considered
each survey (N = 36 dawns), they counted the number of songs in our acoustic analyses. At each sampling site, we performed a
and calls emitted by three or four different males located within single line transect crossing the ARU’s location and using a 160-
a 250 m radius from the observer, from 100 to 10 min before m maximum detection band on each side of the observer. We
sunrise. The authors split the continuous survey into 5-min selected this width given that Dupont’s Lark vocalizations may
intervals and annotated the total number of Dupont’s Lark be heard from long distances (up to 1500 m; Laiolo et al. 2007,
vocalizations (songs and territorial calls) for each interval. To Vögeli et al. 2010). According to a previous assessment of
facilitate comparisons to our acoustic data, we selected surveys counting methods for estimating the number of Dupont’s Lark
carried out during the same period that we performed acoustic males, we assumed a detection probability of 1 for singing males
monitoring (mid-April to mid-June, N = 24 surveys, 83 males) within the 160-m detection band (Pérez-Granados and López-
and only considered the six 5-min surveys carried out between 60 Iborra 2017). Censuses were designed to cross over the recorder
and 30 min before sunrise, which corresponds to the recording location, began 60 min before sunrise, and lasted for
timing at our sampling sites. The probability of detecting a approximately 30 min. The locations of vocalizing males were
Dupont’s Lark vocalization by an observer at distances < 250 m estimated acoustically and recorded by GPS. The number of
has been considered to be 1 (probability of detection = 1 for birds males detected per site refers just to the number of vocalizing
within a 500 m radius of the observer; Pérez-Granados and males detected within the 160 m radius of the recorder. Censuses
López-Iborra 2017). The ACR was estimated by averaging (with were performed one or two dawns after the recorders were
95% confidence interval, 95% CI) the cue rate of each Dupont’s retrieved to avoid modifying the natural vocal behavior of the
Lark male monitored (vocalizations per unit of time per bird). Dupont’s Lark while recording.
The two other methods we used estimate ACR from public or
private (authors’ own data) archival recordings of single males. Statistical analyses
For each methodology, we estimated the ACR (mean ± 95% CI) For each of the 31 sampling sites, we estimated the number of
by averaging the cue rate of all Dupont’s Lark males recorded. Dupont’s Lark males within the considered radius (160 m) by
dividing the total number of Dupont’s Lark cues detected perAvian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
minute in each recording by the average cue rate of the species effort invested in recording five males for 15 min each was 0.06 (5
estimated for each of the three approaches (Lambert and males × 15 min)/(41 males × 30 min). The “effectiveness” was
McDonald 2014). We only considered those cues that reached the estimated as the proportion improvement in ACR accuracy, which
amplitude threshold (144.5 U). was calculated as the percent reduction in the difference in ACR
estimated between the training and test data sets with respect to
We also estimated the minimum and maximum number of males
the maximum distance (i.e., 1.41 ACR difference for 1 male and
per sampling site (rounded to the nearest whole number)
5 min of recording). Finally, the cost-effectiveness was calculated
according to the ACR 95% CIs. In all patches where at least one
as the difference between the cost and the effectiveness values
call was detected, we estimated a minimum of one male. We used
calculated. We used the “RcppAlgos” package (Wood 2020) in R
95% CIs to evaluate whether there were differences (1) among
3.6.2 (R Core Team 2019) to obtain the possible combinations
ACRs estimated using each methodology, and (2) between
from 1 to 40 males.
numbers of males detected using traditional field surveys and
those estimated using each cue counting methodology, at each
site and at the population level (all patches pooled). Differences RESULTS
in ACRs were considered to occur when 95% CIs did not overlap. We detected a total of 4534 Dupont’s Lark vocalizations in the
Likewise, we considered numbers of males detected by traditional recordings collected in the 31 sampling patches (min: 5, max: 521,
field surveys to differ when they were outside the 95% CIs duration: 30 min). The number of males per sampling site detected
estimated by cue counting. Additionally, we estimated the slope by traditional field surveys around ARUs varied from 1 to 5 for
of a no-intercept linear regression (glm function in R 3.6.2) by a total of 65 males counted.
regressing the number of males estimated using traditional field The estimated Dupont’s Lark ACR differed among the
surveys (observed values, y-axis) vs. the estimated number of methodologies, with the exception of the use of directional
males using each of the three approaches (predicted values, x- recordings and behavioral field studies, for which the 95% CI
axis; Piñeiro et al. 2008). We evaluated the model predictions by overlapped (Table 1). The lowest ACR was obtained using data
comparing the slope obtained for each approach to the 1:1 line collected in field studies of the species’ vocal behavior (2.52
(perfect match) and considered the approach with the slope closer vocalizations/min) followed by the use of directional recordings
to 1 (less bias from the 1:1 line) as the most adequate one (Piñeiro (3.09 vocalizations/min). The highest ACR was obtained when
et al. 2008). analyzing the recordings uploaded to the xeno-canto database
To assess the minimum number of males and recording duration (5.76 vocalizations/min; Table 1).
necessary to obtain a reliable population cue rate, we created a
curve of ACR accuracy. For this purpose, we used data collected Table 1. Average cue rate (and 95% CI) of the Dupont’s Lark
by Pérez-Granados et al. (2018c) and curated the data as explained estimated using three methods.
in Methods: Average cue rate. We split the data set in half, each
with 41 individual recordings, to form training and test data sets. Method Individual cue
†
Sites (%) Total males (%)
‡
Each data set comprised males recorded in different years (2013 rate
and 2014), and although all males belonged to the same Behavioral field 2.52 (2.29–2.75) 24 (77.4%) 59–70 (+0.8%)
population, they were recorded at locations separated by > 500 study
m to increase the probability of monitoring different birds among Directional 3.09 (2.47–3.71) 23 (74.2%) 49–68 (−10%)
recordings
years. For each data set, we estimated the ACR for 1 to 40 males,
Xeno-canto 5.76 (4.90–6.62) 17 (54.8%) 36–41 (−40.8%)
using 5 to 30 min of recordings at 5-min intervals. This procedure database
was done for all possible combinations of males when there were †
Number of sampling sites (N = 31) in which the number of males
< 10,000 combinations (1, 2, 39, and 40 males), and for 10,000 detected using traditional field surveys was within the 95% CI of the
random combinations when all possible combinations exceeded number of birds estimated by applying cue counting.
‡
this threshold for computational optimization purposes. We Total number of males (minimum–maximum) estimated for the whole
estimated the difference (in absolute value) in ACR between the area surveyed and, in parentheses, the percentage difference between the
mean number of males estimated by applying cue counting and the
training and test data sets (hereafter ACR accuracy), where lower number of males detected using traditional field surveys (N = 65).
values indicate greater accuracy. This process was repeated 1000
times. We then calculated the mean and SD of ACR accuracy
Accordingly, the numbers of males estimated per sampling site
over all combinations independently for each number of recorded
and at the population level by cue counting differed by
males (1 to 40) and recording durations (from 5 to 30 min at 5-
methodology (Table 2). The best results were obtained when ACR
min intervals). We excluded the combination for 41 males because
was estimated using directional recordings. Using this approach,
it reports a unique value with no variance.
the number of birds estimated around ARUs was considered well
Lastly, we carried out a cost-effectiveness assessment to identify estimated in 23 of the 31 sampling sites (74.2%), and the slope of
the minimum numbers of recordings and timings that are required the linear regression was the one with the least bias from the 1:1
to obtain a reliable ACR while maximizing survey effectiveness. line (slope = 1.03, F1,30 = 160.5, P < 0.0001, adjusted-R2 = 0.84;
The “cost,” or the sampling effort invested in recording a certain Fig. 2B). Total bird numbers were underestimated by 10% when
number of males over a specific duration, was estimated as the compared to the number of males counted using traditional field
proportion of time invested with respect to the maximum time surveys (Table 1). Using the ACR estimated through behavioral
invested under all monitoring scenarios (i.e., 1230 min invested field study, the number of birds was well estimated in 24 sampling
in recording 41 males for 30 min each). For instance, the sampling sites (77.4%), bird numbers were overestimated by 0.8% (Table 1),Avian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
Table 2. Number of Dupont’s Lark vocalizations detected per minute of acoustic monitoring on each sampling site, numbers of males
detected using traditional field surveys, and numbers of males estimated using each cue counting method. The number of males estimated
using each counting method is shown as mean, 95% confidence interval (between brackets), and the minimum and maximum number
of males estimated rounded to the nearest whole number.
Number of males estimated by cue counting
Site Vocalizations Number of Behavioral field study Directional recordings Xeno-canto recordings
(number/min) males
Barcones A 15.30 3 6.1 (5.6–6.7) 6–7 5.0 (4.1–6.2) 4–6 2.7 (2.3–3.1) 2–3
Barcones B 12.90 5 5.1 (4.7–5.6) 5–6 4.2 (3.5–5.2) 3–5 2.2 (1.9–2.6) 2–3
Barcones C 12.91 3 5.1 (4.7–5.6) 5–6 4.2 (3.5–5.2) 3–5 2.2 (1.9–2.6) 2–3
Barcones D 17.37 4 6.9 (6.3–7.6) 6–8 5.6 (4.7–7.0) 5–7 3.0 (2.6–3.5) 3–4
Alcubilla de las Peñas A 0.93 1 0.4 (0.3–0.4) 1–1 0.3 (0.2–0.4) 1–1 0.2 (0.1–0.2) 1–1
Alcubilla de las Peñas B 1.57 1 0.6 (0.6–0.7) 1–1 0.5 (0.4–0.6) 1–1 0.3 (0.2–0.3) 1–1
Alfés A 2.53 1 1.0 (0.9–1.1) 1–1 0.8 (0.7–1.0) 1–1 0.4 (0.4–0.5) 1–1
Alfés B 4.27 3 1.7 (1.6–1.9) 2–2 1.4 (1.1–1.7) 1–2 0.7 (0.6–0.9) 1–1
Alfés C 6.67 3 2.6 (2.4–2.9) 2–3 2.2 (1.8–2.7) 2–3 1.2 (1.0–1.4) 1–1
Atienza 3.47 2 1.4 (1.3–1.5) 1–2 1.1 (0.9–1.4) 1–1 0.6 (0.5–0.7) 1–1
Romanillos de Atienza 0.33 1 0.1 (0.1–0.1) 1–1 0.1 (0.1–0.1) 1–1 0.1 (0.1–0.1) 1–1
Bujalcayado 8.93 3 3.5 (3.2–3.9) 3–4 2.9 (2.4–3.6) 2–4 1.5 (1.3–1.8) 1–2
La Olmeda de Jadraque 7.13 4 2.8 (2.6–3.1) 3–3 2.3 (1.9–2.9) 2–3 1.2 (1.1–1.5) 1–1
Morenilla 4.30 3 1.7 (1.6–1.9) 2–2 1.4 (1.2–1.7) 1–2 0.7 (0.6–0.9) 1–1
El Pobo de Dueña 0.90 1 0.4 (0.3–0.4) 1–1 0.3 (0.2–0.4) 1–1 0.2 (0.1–0.2) 1–1
Luzón 5.60 2 2.2 (2.0–2.4) 2–2 1.8 (1.5–2.3) 2–2 1.0 (0.8–1.1) 1–1
Retortillo 1.07 1 0.4 (0.4–0.5) 1–1 0.3 (0.3–0.4) 1–1 0.2 (0.2–0.2) 1–1
Barahona A 3.60 2 1.4 (1.3–1.6) 1–2 1.2 (1.0–1.5) 1–1 0.6 (0.5–0.7) 1–1
Barahona B 0.60 1 0.2 (0.2–0.3) 1–1 0.2 (0.2–0.2) 1–1 0.1 (0.1–0.1) 1–1
Barahona C 1.60 1 0.6 (0.6–0.7) 1–1 0.5 (0.4–0.7) 1–1 0.3 (0.2–0.3) 1–1
Barahona D 2.07 1 0.8 (0.8–0.9) 1–1 0.7 (0.6–0.8) 1–1 0.4 (0.3–0.4) 1–1
Rello A 5.47 3 2.2 (2.0–2.4) 2–2 1.8 (1.5–2.2) 1–2 1.0 (0.8–1.1) 1–1
Rello B 1.47 1 0.6 (0.5–0.6) 1–1 0.5 (0.4–0.6) 1–1 0.3 (0.2–0.3) 1–1
Rello C 6.73 3 2.7 (2.4–2.9) 2–3 2.2 (1.9–2.8) 2–3 1.2 (1.0–1.4) 1–1
Rello D 6.90 3 2.7 (2.5–3.0) 3–3 2.2 (1.9–2.8) 2–3 1.2 (1.0–1.4) 1–1
Carboneras de Guadazaón 0.17 1 0.1 (0.1–0.1) 1–1 0.1 (0.1–0.1) 1–1 0.1 (0.1–0.1) 1–1
Olmedillas 2.23 1 0.9 (0.8–1.0) 1–1 0.7 (0.6–0.9) 1–1 0.4 (0.3–0.5) 1–1
Moya-Huertos 1.27 1 0.5 (0.5–0.6) 1–1 0.4 (0.3–0.5) 1–1 0.2 (0.2–0.3) 1–1
Rayana 4.73 2 1.9 (1.7–2.1) 2–2 1.5 (1.3–1.9) 1–2 0.8 (0.7–1.0) 1–1
Retortillo B 1.93 1 0.8 (0.7–0.8) 1–1 0.6 (0.5–0.8) 1–1 0.3 (0.3–0.4) 1–1
Retortillo C 6.20 3 2.5 (2.3–2.7) 2–3 2.0 (1.7–2.5) 2–3 1.1 (0.9–1.3) 1–1
Total 65 59–70 49–68 36–41
and the slope of the linear regression was 0.84 (F1,30 = 159.4, P < longer recordings (Fig. 3B). However, with ≥ 5 males, the cost-
0.0001, adjusted-R2 = 0.84; Fig. 2A). When using the ACR effectiveness of short recordings (5 min) is greater, and sampling
obtained from recordings uploaded to xeno-canto, the number of effort should be invested in recording more males over shorter
sites in which bird numbers were well estimated decreased to 17 durations (Fig. 3B). Taking everything into consideration,
(54.8% of cases), and bird densities were underestimated by 41%. recording more males should be prioritized, rather than longer
The slope of the linear regression was very different from 1:1 recordings.
(slope = 1.91, F1,30 = 160.5, P < 0.0001, adjusted-R2 = 0.84; Fig.
2C). DISCUSSION
The distance between the ACR calculated for the training and Here, we tested the effectiveness of cue counting for estimating
test data sets decreased as the number of males recorded and bird abundance using ARUs. We found that cue counting can be
recording duration increased (Fig. 3A). Recording duration had useful to infer Dupont’s Lark numbers around ARUs, in agreement
a greater effect on ACR accuracy when recording a few males (< with prior studies using ARUs to estimate cetacean (e.g., Hiby and
3–4 males), and its effect was decreased when recording a greater Ward 1986, Marques et al. 2009, 2011) and bird abundance
number of males (Fig. 3A). The cost-effectiveness assessment (Lambert and McDonald 2014, Sebastián-González et al. 2018).
showed that surveys that aim to estimate a reliable ACR are Our results suggest that cue counting could be a reliable and easy
maximized at different points depending on the number of males solution to provide a concrete number of birds singing around
and duration monitored (Fig. 3B). For example, the cost- ARUs while avoiding the need to collect additional data during
effectiveness of making 25-min recordings reaches a maximum at recording, as most other methods require (e.g., Oppel et al. 2014,
five males, whereas it is maximized at 20 males for 5-min Van Wilgenburg et al. 2017, Yip et al. 2020). However, our results
recordings (Fig. 3B). With 1–3 males, long recordings are more also show that cue counting requires the estimate of a reliable ACR
cost-efficient, and thus, sampling effort should be invested in for the monitored species because population estimates differ
greatly with the ACR applied. Therefore, cue counting should beAvian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
Fig. 2. Regression scatterplot showing Dupont’s Lark male Fig. 3. (A) Average cue rate (ACR) accuracy for combinations
abundance observed using traditional field surveys vs. of number of males recorded and recording duration. ACR
estimated using cue counting and behavioral field study (A), accuracy was calculated as the difference (absolute value)
directional recordings (B), and xeno-canto recordings (C). Solid between the ACR for the training and test data sets; lower
line = no-intercept linear regression, grey bands = 95% values indicate greater accuracy. (B) Cost-effectiveness of
confidence intervals, dashed line = 1:1 line. surveys to estimate the ACR of the Dupont’s Lark for
combinations of number of males recorded and recording
duration. Solid line = mean, colored bands = standard
deviation, vertical dashed lines = number of males at which the
cost-effectiveness is maximized in each scenario (also provided
within parentheses in the legend).
used only after an accurate and reliable estimation of the species’
ACR is obtained. We next discuss some potential problems and
drawbacks when estimating the Dupont’s Lark’s ACR. It is
beyond our scope to discuss the accuracy of the estimation of the
number of males because that was obtained simply by dividing
the number of vocalizations per recording (which did not vary
among methods) by the ACR estimated using each approach.
Although the 95% CI of the ACR estimated using directional
recordings and behavioral field data overlapped, the ACR
estimated using directional recordings showed less bias from the
1:1 line. The ACR estimated through behavioral field study
provided more biased predictions and was also a more effort-
it does not appear to be useful for estimating the ACR of the
intensive method. A previous study showed that cue rate of
Dupont’s Lark. Recordings uploaded to xeno-canto may not
Dupont’s Lark males may differ by up to 50% between close
represent a random sample of the species’ vocal behavior, with
populations (< 5 km; Pérez-Granados and López-Iborra 2020),
quieter birds being underrepresented or not recorded (Pagani-
suggesting that differences found among approaches might be
Núñez et al. 2018). However, monitoring birds with low vocal
related to variations in the ACR among Dupont’s Lark
activity, and even nonvocalizing birds, may help to provide a more
populations. Directional recordings were collected in the study
representative estimate of the ACR of a target species (Sebastián-
area, whereas the behavioral field study was performed in a
González et al. 2018). Likewise, recordists may have uploaded a
different population located 140 km distant, which suggests that
short fraction of the original recording to provide a good and
cue counting may perform better when using an ACR estimated
clear example of the vocal activity of the species, which may
for the monitored population.
explain why a large number of the recordings we analyzed were
The actual number of birds was underestimated by approximately very short. The short duration of the recordings uploaded to
41% when applying the ACR estimated using the xeno-canto xeno-canto is considered one of the main drawbacks of the
database. Many previous studies have demonstrated the utility of database (Vellinga and Planqué 2015). Although we removed
xeno-canto data for acoustic research (e.g., Weir and Wheatcroft recordings with more than one bird vocalizing from the analysis,
2011, Benedetti et al. 2018, Pagani-Núñez et al. 2018). However, we cannot rule out the possibility that more than one bird mayAvian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
have been recorded in some of the recordings analyzed. The Miliaria calandra) can decrease their vocal activity under high
combination of these factors may contribute to explaining the density because they dedicate a large proportion of time to
high ACR estimated using xeno-canto recordings compared to territorial defence and listening to rivals (Olinkiewicz and Osiejuk
those from directional recordings and behavioral field data. We 2003). For these species, the number of birds estimated at high
did not control for time of day or season when selecting recordings density may be underestimated.
from the xeno-canto database. It is possible that future studies
Several factors may alter the cue rate of birds, such as breeding
working with birds with a greater number of uploaded recordings
status (Amrhein et al. 2002), weather conditions (Robbins 1981),
may remove those obtained during the nonbreeding season or
moon phase (York et al. 2014), and time of day and season
outside the period of interest and therefore will likely obtain a
(Amrhein et al. 2002, York et al. 2014). For example, the cue rate
more reliable ACR for the target species (Buckland 2006).
of the Dupont’s Lark increases as the breeding season progresses
The measurement of a reliable individual cue rate is needed to (Laiolo and Tella 2008, Pérez-Granados et al. 2018c), is lower
obtain an unbiased ACR for the target species and might be also when the moon is full (Pérez-Granados and López-Iborra 2020),
useful for other purposes such as predicting male breeding status and is reduced when there are few neighboring males (Laiolo et
(Upham-Mills et al. 2020). However, there is very limited al. 2008). These findings suggest that it might be necessary to
information about the number of individuals and duration that monitor a representative number of males vocalizing under
each individual needs to be monitored to estimate a reliable ACR. different environmental and phenological conditions to obtain a
Our cost-effectiveness assessment showed that surveys to estimate reliable ACR. Among the sample of birds monitored, it is also
the cue rate of the Dupont’s Lark are maximized at different interesting to include nonvocalizing individuals (Sebastián-
points, depending on the number of males and duration of González et al. 2018). It is important to be sure that the silent or
monitoring. In accordance with our results, if surveys are less vocally active behavior of monitored individuals is part of
conducted at small populations, sampling effort should be the natural vocal behavior of the target species and not influenced
invested in recording over longer durations (up to 30 min; Fig. by an observer’s presence.
3B) because of the positive effect of time on ACR accuracy with
Buckland (2006) suggested that cue counting works best when
≤ 3 males (Fig. 3A). However, with larger populations, recording
cues are short and well defined, and that for species with long
more males for shorter durations is preferable. In the latter
songbursts, the start of the songburst might be defined as the cue.
scenario, the effect of recording duration on ACR accuracy is
We used a similar approach in which we used the ending and
decreased, so longer recordings are less cost-efficient. In an ideal
definite “whee-ur-wheeee” of the Dupont’s Lark song, usually
scenario with unlimited males, time, and budget, the most
composed of 4–5 song types (Laiolo et al. 2008, Pérez-Granados
accurate ACR is obtained when recording 20 males for 5-min
et al. 2016), to count the number of songs within a recording. Our
durations, while maximizing surveys effectiveness. Thus,
approach also agrees with the recommendation made by
recording more males should be prioritized whenever possible,
Buckland (2006) that for species with a wide range of
rather than longer recordings. Our results may not be extrapolated
vocalizations, only those easily detectable might be considered as
directly to other bird species, but we hope that they will be useful
a cue (Buckland 2006). The estimation of the cue rate should be
for future studies aiming to estimate the cue rate of other bird
conducted at the same time and place as the main surveys
species, as well as for studies applying cue counting for bird density
(Buckland et al. 2008) or substantial bias could appear. For
estimation. The need to monitor the vocal behavior of a
example, previous studies of the Dupont’s Lark state that the cue
representative sample of birds to estimate an unbiased ACR has
rate may differ by up to 50% among close populations (Pérez-
been described as one of the main disadvantages of cue counting
Granados and López-Iborra 2020). The vocal activity of birds,
(Buckland et al. 20008).
including the Dupont’s Lark (Pérez-Granados et al. 2018c),
Our study was carried out at sites with low bird densities, and we usually reaches a maximum during dawn (Gil and Llusia 2020).
expect that the method may be less effective or more biased for Therefore, for most species, it would be desirable to perform
high-density populations. For example, the relationship between surveys and to estimate the ACR during dawn. The performance
recorded vocal activity rate and Dupont’s Lark abundance around of cue counting, when using ARUs, may be higher when working
recorders shows a logarithmic curve (Pérez-Granados et al. with species that have a constant cue rate among individuals
2019b), which suggests that the effectiveness of cue counting for (Lambert and McDonald 2014) and with territorial birds that
estimating Dupont’s Lark abundance may be decreased when undertake few movements during the singing period, but further
applied at high densities. Nonetheless, the Dupont’s Lark is a research should evaluate the effect of bird movements on the
threatened species (Gómez-Catasús et al. 2018), listed as estimated cue rates. The necessary number of males and duration
“Vulnerable” in Europe by the International Union for the of monitoring may be adjusted according to population size. In
Conservation of Nature (BirdLife International 2015). Therefore, summary, a good strategy to obtain accurate results using cue
the species is usually distributed at low densities (mean of 0.30 counting would be as follows: (1) use directional microphones to
males/10 ha in Spain; Suárez 2010), and cue counting may be a monitor the cue rate of a representative sample of individuals
good approach to provide a reliable range of males around comprising a minimum of ~10 males, (2) perform subsequent
recorders because their density usually ranges between one and surveys within the same geographic areas where ACR estimation
three (as in the present study). Nonetheless, the method’s is conducted, (3) monitor individuals for longer periods (25 min)
effectiveness for estimating bird abundance should be evaluated if fewer than 5 males are available or for 5 min if 20 males are
at high density and when studying different groups of birds such available, (4) include nonvocalizing birds, and (5) perform
as those living in colonies or singing while flying (e.g., Arneill et monitoring over several days and under a variety of weather
al. 2020). For example, some passerines (e.g., Corn Bunting conditions. The same reasoning should be valid for obtaining anAvian Conservation and Ecology 16(1): 11
http://www.ace-eco.org/vol16/iss1/art11/
accurate number of vocalizations per sampling site through
ARUs. Indeed, previous studies with three bird species showed Acknowledgments:
that programming ARUs to record for thirteen consecutive days
This study was supported by the Programa de Investigación y
rather than recording for a single day, as we did in our study, may
Conservación del Zoo de Barcelona within the project “Nuevas
decrease the coefficient of variation of the vocal activity rate at
tecnologías para viejos trabajos: Uso de grabadores automáticos
a population level from 90–100% to 10% (Pérez-Granados et al.
para la detección y censo de especies raras y amenazadas: El caso
2019c, Pérez-Granados and Schuchmann 2020). Therefore, it is
de la alondra ricotí en Lleida y otras poblaciones pequeñas”; the
likely that the performance of our method would have been higher
Education, Youth and Sport Bureau (Madrid Regional
if we had recorded over several days.
Government); and the European Social Fund for the Youth
Our study demonstrates the ability of passive acoustic Employment Initiative (PEJ15/AMB/AI-0059 and PEJ1D-2018-
monitoring, and more specifically of cue counting, to estimate PRE/AMB-8063). This work is a contribution to the Excellence
bird abundance reliably around ARUs, which is in agreement with Network Remedinal 3CM (S2013/MAE2719), supported by
several recent studies that showed the utility of automated Comunidad de Madrid; and to the projects “Census of Dupont’s
recorders to infer bird numbers (e.g., Oppel et al. 2014, Van Lark in Guadalajara 2017 (SSCC/046/2017)”, supported by Junta
Wilgenburg et al. 2017, Darras et al. 2018, Sebastián-González de Comunidades de Castilla-La Mancha; LIFE Ricotí (LIFE15-
et al. 2018, Bombaci and Pejchar 2019, Pérez-Granados et al. NAT-ES-000802), supported by the European Commission, and
2019c, Yip et al. 2020; reviewed by Pérez-Granados and Traba “BBVA-Dron Ricotí”, funded by the BBVA Foundation. We thank
2021). Our study also provides valuable data about the numbers Gerard Bota, David Giralt, and Josep Albarracín for their close
of individuals and monitoring durations needed to maximize collaboration and help. We also thank Alexander Bond, Steven Van
surveys that aim to estimate a reliable ACR for a target species, Wilgenburg, and two anonymous reviewers whose comments helped
which is paramount before applying cue counting to estimate bird to improve the manuscript.
abundance. Otherwise, cue counting may lead to inadequate
population size estimates. The development of acoustic-derived
indices has opened the door to use ARUs to track bird population LITERATURE CITED
changes over time (e.g., Buxton and Jones 2012). In this respect, Amrhein, V., P. Korner, and M. Naguib. 2002. Nocturnal and
cue counting may be useful for assessing bird population changes diurnal singing activity in the nightingale: correlations with
using archived recordings because our method can be applied mating status and breeding cycle. Animal Behaviour 64
without the collection of additional data during recording, given (6):939-944. https://doi.org/10.1006/anbe.2002.1974
that we estimated the ACR using previously collected data.
Nonetheless, it is necessary to include a measurement of the Arneill, G. E., E. J. Critchley, S. Wischnewski, M. J. Jessopp, and
effective detection radius of the recorder (see Buckland et al. 2011) J. L. Quinn. 2020. Acoustic activity across a seabird colony reflects
or use a fixed radius around ARUs (Lambert and McDonald patterns of within-colony flight rather than nest density. Ibis 162
2014) to obtain density estimates. The ACR may differ over time (2):416-428. https://doi.org/10.1111/ibi.12740
depending on exogenous (e.g., time of day, weather conditions) Benedetti, Y., K. Slezak, A. P. Møller, F. Morelli, and P.
and endogenous factors (e.g., breeding status; Amrhein et al. Tryjanowski. 2018. Number of syllables in cuckoo Cuculus
2002, Upham-Mills et al. 2020); thus, the utility of cue counting canorus calls: a test using a citizen science project. Scientific
for estimating bird densities using archived recordings may be Reports 8:12872. https://doi.org/10.1038/s41598-018-31329-1
limited and requires further research. Likewise, future studies
should assess how the ACR changes over time and among BirdLife International. 2015. Chersophilus duponti. The IUCN
populations, which may provide useful insights about the Red List of Threatened Species 2015: e.T22717380A60256867.
effectiveness of cue counting for estimating bird abundance at Birdlife International, Cambridge, UK. [online] URL: https://
large spatial and temporal scales. Future works using automated www.iucnredlist.org/species/22717380/60256867
signal recognition software for scanning the recordings should Bombaci, S. P., and L. Pejchar. 2019. Using paired acoustic
estimate and take into account the recall rate of the recognizer sampling to enhance population monitoring of New Zealand’s
(Knight et al. 2017, Sebastián-González et al. 2018) when forest birds. New Zealand Journal of Ecology 43(1):3356. https://
estimating the number of bird vocalizations in a sound recording. dx.doi.org/10.20417/nzjecol.43.9
We encourage researchers to evaluate the effectiveness of cue
counting for more bird species because the current knowledge Buckland, S. T. 2006. Point-transect surveys for songbirds: robust
about the species and conditions in which this method can be methodologies. Auk 123(2):345-357. https://doi.org/10.1093/
applied is still limited (Lambert and McDonald 2014, Sebastián- auk/123.2.345
González et al. 2018). Buckland, S. T., D. R. Anderson, K. P. Burnham, J. L. Laake, D.
L. Borchers, and L. Thomas. 2001. Introduction to distance
sampling: estimating abundance of biological populations. Oxford
Responses to this article can be read online at: University Press, Oxford, UK.
https://www.ace-eco.org/issues/responses.php/1801
Buckland, S. T., S. J. Marsden, and R. E. Green. 2008. Estimating
bird abundance: making methods work. Bird Conservation
International 18(S1):S91-S108. https://doi.org/10.1017/S0959270908000294You can also read
